国际寡占竞争框架下的最优战略研发政策分析

【作者】 陈磊；

【导师】 袁志刚；

【作者基本信息】 复旦大学 ， 西方经济学， 2006， 博士

【摘要】 本文所要回答的问题有四个：①在一个离散框架下，如果关于企业或政府预先承诺能力的不同，学习效应与外溢效应的大小对最优研发政策到底会造成什么影响?②较高的政府研发补贴真的对一个知识密集型幼稚产业的成长是必要的吗?③如何解释中国知识政策现实—近期我国从主要依赖海外技术外溢的“技术换市场”的政策开始转向于“自主创新”政策；④如何为一个正在发展中的战略产业设置一条最优研发补贴路径?本文第二章回顾了相关文献；第三章，根据企业与政府预先承诺能力的不同，在一个具有研发中学习效应的动态框架下考察了三种情形下的最优研发政策：FCE(完全预先承诺均衡)情形下，仅为纯粹转移租金激励；GCE(仅政府预先承诺均)情形下，增加了抵消国内公司战略行为的研发课税激励；SE(序贯均衡)情形下，进一步增加了抵消国内公司操纵政府第二时期补贴战略行为的研发课税激励；在学习效应的较显著情形下，SE有可能削弱通常的幼稚产业保护论调。本文第四章，对称的在一个研发活动中存在外溢效应的动态框架下考察了三种情形下的最优研发政策：FCE(完全预先承诺均衡)情形下，存在转移租金激励、外溢及反外溢激励的权衡，且在外溢及反外溢激励较大的情形下，最优政策表现为研发课税；GCE(仅政府预先承诺均)情形下，除了前三种激励，还增加了一项抵消企业策略行为的补贴激励项；SE(序贯均衡)情形下，进一步增加了抵消国内公司操纵第二时期政府补贴策略行为的补贴激励。在具有显著外溢效应的情形下，SE有可能削弱通常的“市场换技术”论调。本文第五章，在一个连续技术创新框架下考察最优研发补贴路径，就补贴时间路径的差异而言：具有学习效应的补贴强度最大，基本模型次之，外溢效应模型补贴强度最弱。第六章是本文的结论，并对开头的四个问题作出了回答。更多还原

【Abstract】 In this paper, we try to answer: (1) if learning by doing, spillover effects exist in R&D activities, what kinds of incentive components emerge in an optimal strategic R&D policy, and how these components changes with these two dynamic linkage? (2) Is the high R&D subsidies necessary for the development knowledge-intensive infant industry? (3)How to give an explanation for this phenomenon: China government is changing its technology policy from "Market exchanging for know-how" to "technology self-innovation"; (4) How to set an optimal R&D subsidy path for a developing strategic industry?In Chapter 2, we review all the relevant literatures; in Chapter 3, according to the three different assumptions about pre-commitment of government and enterprise, we examines the implications for optimal R&D policies in a two-period (four -stage) Cournot oligopoly game with learning by doing in R&D activities: in FCE, optimal subsidies only come from incentive of rent shifting; in GCE, optimal subsidies have an extra taxation term which offsets the strategic behavior by the home firm; in SE, optimal subsidies further includes another extra taxation term which offsets the strategic behavior by the home firm aimed at manipulating period-2 subsidy. Provided learning by doing in R&D activities is significant, results in SE will possibly weaken the infant-industry argument; In Chapter 4, we symmetrically examine the implications for optimal R&D policies in a two-period (four-stage) Cournot oligopoly game with spillover effects in R&D activities: in FCE, optimal subsidies only come from the trade-off between the incentive of rent shifting and that of spillover and anti-spillover; in GCE, optimal subsidies have an extra subsidy term which offsets the strategic behavior by the home firm; in SE, optimal subsidies further includes another extra subsidy term which offsets the strategic behavior by the home firm aimed at manipulating period-2 subsidy. Provided spillover effect in R&D activities is significant, results in SE will weaken the "Market exchanging for know-how " argument; In chapter 5, under a continuous time framework, we find the difference of subsidy path caused by two dynamic linkage: the higher is the optimal R&D subsidizing path, the stronger is the learning effects in R&D activities; the lower is the optimal R&D subsidizing path, the stronger is the spillover effects in R&D activities. Chapter 6 is our conclusion and the answer to the initial four questions.更多还原